Polycystic ovary syndrome (PCOS) affects 5-10% of women and may be viewed as the combination of hyperandrogenism with the classical features of the metabolic syndrome in young women. PCOS presents a unique opportunity to dissect the relationship between metabolic and cardiovascular risk and sleep disordered breathing (SDB) in a population where intrinsic effects of aging have not yet developed. Because a relationship between obstructive sleep apnea, insulin resistance and elevated testosterone levels has also been observed in men and in women without PCOS, insights gained from studies in PCOS will have broad implications. the Specific Aims of the present application are: Specific Aim 1: to test the hypothesis that sleep disturbances are caused by hyperandrogenemia and hyperinsulinemia that characterize PCOS. Following a detailed baseline evaluation of sleep, hormonal, metabolic and cardiovascular parameters, women with PCOS will be randomized to an 8-week treatment phase with pioglitazone or depot leuprolide plus estrogerdprogestin replacement or placebo. Pioglitazone will reduce insulin levels, and consequently androgen levels, in PCOS. We will compare the effects of androgen reduction alone (depot leuprolide plus estrogerdprogestin) to those of insulin plus androgen reduction achieved with pioglitazone. Primary comparisons will be the change in sleep parameters from baseline between: placebo & pioglitazone; placebo & leuprolide/estrogen/progestin; pioglitazone & leuprolide/estrogen/progestin. Specific Aim 2: to test the hypothesis that sleep disturbances cause the hormonal, metabolic and cardiovascular alterations seen in women with PCOS. PCOS women with SDB and matched control women with SDB will be evaluated at baseline and following 8 weeks of CPAP treatment. The primary comparison will be between baseline and post-treatment parameters in PCOS women. The secondary comparison will be the post-treatment change from baseline between PCOS and control women to test the hypothesis that for the same degree in improvement in SDB, the magnitude of change in metabolic and cardiovascular measures will be greater in PCOS than in controls. Specific Aim 3: to test the hypothesis that in normal young women, experimental manipulation of sleep that recapitulates the sleep disturbances characteristic of women with PCOS will result in metabolic, hormonal, and cardiovascular alterations that are typical of the metabolic syndrome. A group of healthy young women will be studied twice using a randomized cross-over design. In one study, REM sleep will be fragmented by experimentally induced microarousals for 3 consecutive nights and non- REM sleep will be left undisturbed. In the other, slow wave activity will be suppressed without awakening the subject and REM sleep will be left undisturbed. Each study will be preceded by 2 nights of basetine sleep.